Adam A. Gronowski
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Adam A. Gronowski has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Ana Lúcia Amaral Eisenberg
Division of Pathology, Brazilian National Cancer Institute (INCA), Rio de Janeiro, Brazil
1 shared publication - Bahman Gharesifard
Department of Mathematics and Statistics, Queens University, Kingston, ON, Canada
1 shared publication - Fady Alajaji
Department of Mathematics and Statistics, Queens University, Kingston, ON, Canada
1 shared publication - Guang-Peng Wu
MOE Key Laboratory of Macromolecular Synthesis and Functionalization Zhejiang Key Laboratory of Advanced Organic Materials and Technologies Department of Polymer Science and Engineering Zhejiang University Hangzhou 310058 China
1 shared publication H. L. Yeager
1 shared publicationJ.-S. Kim
1 shared publication
Latest Publications
- 5 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 19 (15566013) , 1630-1645
Classification Utility, Fairness, and Compactness via Tunable Information Bottleneck and Rényi Measures
Designing machine learning algorithms that are accurate yet fair, not discriminating based on any sensitive attribute, is of paramount importance for society to accept AI for critical applications. In this article, we propose a novel fair representation learning method termed the Rényi Fair Information Bottleneck Method (RFIB) which incorporates constraints for utility, fairness, and compactness (compression) of representation, and apply it to image and tabular data classification. A key attribute of our approach is that we consider - in contrast to most prior work - both demographic parity and equalized odds as fairness constraints, allowing for a more nuanced satisfaction of both criteria. Leveraging a variational approach, we show that our objectives yield a loss function involving classical Information Bottleneck (IB) measures and establish an upper bound in terms of two Rényi measures of order $ \boldsymbol {\alpha }$ on the mutual information IB term measuring compactness between the input and its encoded embedding. We study the influence of the $ \boldsymbol {\alpha }$ parameter as well as two other tunable IB parameters on achieving utility/fairness trade-off goals, and show that the $ \boldsymbol {\alpha }$ parameter gives an additional degree of freedom that can be used to control the compactness of the representation. Experimenting on three different image datasets (EyePACS, CelebA, and FairFace) and two tabular datasets (Adult and COMPAS), using both binary and categorical sensitive attributes, we show that on various utility, fairness, and compound utility/fairness metrics RFIB outperforms current state-of-the-art approaches.
- 5 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 19 (15566013) , 1630-1645
Classification Utility, Fairness, and Compactness via Tunable Information Bottleneck and Rényi Measures
Designing machine learning algorithms that are accurate yet fair, not discriminating based on any sensitive attribute, is of paramount importance for society to accept AI for critical applications. In this article, we propose a novel fair representation learning method termed the Rényi Fair Information Bottleneck Method (RFIB) which incorporates constraints for utility, fairness, and compactness (compression) of representation, and apply it to image and tabular data classification. A key attribute of our approach is that we consider - in contrast to most prior work - both demographic parity and equalized odds as fairness constraints, allowing for a more nuanced satisfaction of both criteria. Leveraging a variational approach, we show that our objectives yield a loss function involving classical Information Bottleneck (IB) measures and establish an upper bound in terms of two Rényi measures of order $ \boldsymbol {\alpha }$ on the mutual information IB term measuring compactness between the input and its encoded embedding. We study the influence of the $ \boldsymbol {\alpha }$ parameter as well as two other tunable IB parameters on achieving utility/fairness trade-off goals, and show that the $ \boldsymbol {\alpha }$ parameter gives an additional degree of freedom that can be used to control the compactness of the representation. Experimenting on three different image datasets (EyePACS, CelebA, and FairFace) and two tabular datasets (Adult and COMPAS), using both binary and categorical sensitive attributes, we show that on various utility, fairness, and compound utility/fairness metrics RFIB outperforms current state-of-the-art approaches.
- 21 January 2003
- Published by Wiley in Journal of Applied Polymer Science
- Vol. 87 (14) , 2360-2364
Synthesis of butyl rubber in hexane using a mixture of Et2AlCl and EtAlCl2 in the initiating system
- 21 January 2003
- Published by Wiley in Journal of Applied Polymer Science
- Vol. 87 (14) , 2360-2364
Synthesis of butyl rubber in hexane using a mixture of Et2AlCl and EtAlCl2 in the initiating system
- 1 October 1994
- Published by American Chemical Society (ACS) in Macromolecules
- Vol. 27 (22) , 6541-6550
Structures and Properties of Hydrocarbon Ionomer Membranes. 2. Polystyrene-Based Cation Exchangers
- 1 October 1994
- Published by American Chemical Society (ACS) in Macromolecules
- Vol. 27 (22) , 6541-6550
Structures and Properties of Hydrocarbon Ionomer Membranes. 2. Polystyrene-Based Cation Exchangers